15,16-Dihydrotanshinone I, a novel β-catenin-targeting inhibitor that inhibits its nuclear translocation and reduces downstream CD36 expression in cancer

Abstract

Background

Wnt/β-catenin signaling drives many cancers; however, current inhibition strategies have limitations. Direct β-catenin inhibitors that block nuclear translocation represent an alternative therapeutic approach. We used colorectal cancer (CRC) as a study model, over 90% of the CRC cases harbor mutations in this pathway. Here, we aim to explore the potential of 15,16-dihydrotanshinone-I (DHTS) as a β-catenin inhibitor in cancer.

Methods

Molecular docking, biolayer interferometry assay and cellular thermal shift assay were used to examine the binding between DHTS and β-catenin. Site-directed mutagenesis, in vitro and in vivo studies were used to examine the anti-CRC effects of DHTS following its binding to β-catenin.

Results

We demonstrate, for the first time, that DHTS significantly reduces nuclear β-catenin levels and transcriptional activity in CRC without affecting β-catenin protein stability or conformation. DHTS binds to β-catenin at Ser411. A point mutation at Ser411 disrupts this binding and abolishes the ability of DHTS to suppress nuclear β-catenin, indicating that Ser411 binding is critical for blocking β-catenin nuclear translocation. Furthermore, we identify CD36, a transmembrane fatty acid translocase, as a downstream β-catenin target. DHTS reduces CD36 expression and cellular ATP levels in CRC cells; the reductions are reversed by β-catenin overexpression or stabilization. In CRC-bearing mouse model, DHTS-loaded PLGA-PEG nanoparticles significantly reduce nuclear β-catenin and CD36 expressions in tumors and inhibit tumor growth, consistent with the in vitro findings.

Conclusions

Our study not only reveals the importance of Ser411 in β-catenin function but also paves the path for developing DHTS as a β-catenin inhibitor for CRC therapy.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-025-07317-1.

Introduction

The Wnt/β-catenin signaling pathway plays a role in the development and progression of many cancers. Among these cancers, over 90% of CRC have mutations that activate the Wnt/β-catenin pathway [1–3]. Loss or inactivation of adenomatous polyposis coli results in the constitutive activation of the Wnt/β-catenin signaling pathway, with β-catenin acting as the key effector [4]. Cytoplasmic accumulation of β-catenin leads to its translocation to the nucleus, where it interacts with T cell factor-4 (TCF4) to drive the transcription of target genes that promote CRC development. Studies have shown a correlation between the overexpression of nuclear β-catenin and the occurrence of distant metastasis, lymph node metastasis, and poor prognosis in CRC patients [2, 5]. Elevated nuclear β-catenin expression is significantly associated with reduced disease-free survival (DFS), cancer-specific survival (CSS) and overall survival (OS) in the patients [2]. Indeed, among the various oncogenic pathways involved in the initiation and progression of CRC, the Wnt/β-catenin signaling pathway is the most prominent.

Given CRC is one of the most commonly diagnosed cancers worldwide with increasing prevalence [6–8], various strategies have been explored to inhibit the Wnt/β-catenin signaling pathway in CRC, including targeting the cell membrane to block Wnt ligands, reducing β-catenin stability in the cytoplasm, and disrupting the β-catenin and TCF4 interaction in the nucleus. However, clinical trials of the Wnt/β-catenin inhibitors have shown limited benefits in terms of response rates, survival, and quality of life. Therefore, the Wnt/β-catenin pathway has not been targeted in the clinical setting for caner management [9]. One reason for this limited efficacy is that many proposed Wnt inhibitors that target upstream components of the pathway inadvertently enhance β-catenin phosphorylation, a process often defective in cancer cells [10]. Besides, disrupting the β-catenin and TCF4 interaction with a small molecule inhibitor is challenging because their binding is strong, characterized by a low nanomolar dissociation constant and a large interaction surface area [11, 12]. Due to the substantial accumulation of β-catenin in CRC cells with active Wnt signaling, direct targeting β-catenin is likely to offer greater clinical benefit than targeting other components of the pathway. However, to date, β-catenin has not been successfully targeted [10]. The ineffectiveness of current pathway inhibitors underscores the urgency of developing effective β-catenin-targeted therapies.

15,16-Dihydrotanshinone I (DHTS) is an abietane diterpenoid found in the traditional Chinese medicinal plant Salvia miltiorrhiza Bunge. The plant has been used in traditional Chinese medicine to treat various conditions, including cardiovascular diseases such as hypertension. Modern pharmacology studies have shown that the bioactive compound DHTS is effective against ischemic stroke [13–15]. In traditional Korean medicine, DHTS has also been recognized for its therapeutic effect in treating heart and liver diseases [16]. Previously, a study demonstrated that DHTS binds to the RNA-binding protein human antigen R (HuR), but it displaces only those HuR-bound mRNAs whose affinity for HuR is lower than that of DHTS [17]. Besides its binding to HuR, no other protein targets of DHTS have been reported.

In this study, we evaluated the potential of DHTS as a β-catenin inhibitor in CRC. Our findings support the development of DHTS as an effective β-catenin-targeted therapy for CRC.

Results

DHTS significantly reduces nuclear β-catenin expression and transcriptional activity in CRC cells

Our data indicated that DHTS was a compound that had a potent inhibitory effect on CRC cell growth (Fig. S1). The IC₅₀ values ranged from 0.8 ± 0.054 µM to 1.97 ± 0.342 µM across eight different human CRC cell lines, including HCT116, DLD-1, HT29, SW620, SW480, HCT15, HT55 and Caco-2, as well as mouse CRC cell line CT26 (Table S1). Interestingly, we found that DHTS treatments significantly reduced nuclear β-catenin expressions as demonstrated in CRC cells such as DLD-1 cells (Fig. 1A) and HT55 cells (Fig. 1B). Confocal imaging also showed that the nuclear β-catenin expression in the DLD-1 cells was significantly reduced following DHTS treatments (Fig. S2). The treatments did not affect cytoplasmic β-catenin expressions (Fig. 1A and B). T cell factor 4 (TCF4) and lymphoid enhancer-binding factor-1 (LEF1) are two important transcription factors that interact with β-catenin in the nucleus and initiate gene transcription. A study reported that β-catenin directly activated TCF4 gene transcription [18], suggesting a correlation between the nuclear expressions of β-catenin and TCF4. In our study, we found that inhibiting GSK3β activity and stabilizing free β-catenin levels in CRC by overexpressing β-catenin [19]or treating the cells with lithium chloride (LiCl) significantly increased not only TCF4 but also LEF1 nuclear expressions (Fig. 1C and D). If TCF4 and LEF1 expressions depend on β-catenin, then the reduction of β-catenin following DHTS treatments should also lead to a decrease in their expressions. Indeed, DHTS treatments significantly reduced the expressions of TCF4 and LEF1 in DLD-1 cells (Fig. 1E) and HT55 cells (Fig. 1F). Overexpression of β-catenin abolished the DHTS-reduced TCF4 and LEF1 expressions in these cells (Fig. 1G and H). With Tcf-reporter used in TOP/FLOP flash reporter assay, we found that DHTS treatments significantly reduced the transcription activity of β-catenin/TCF4 complex in the CRC cells (Fig. 1I and J). Our data strongly suggests that DHTS treatment reduces nuclear β-catenin expression and, consequently, the expressions of TCF4 and LEF1, leading to a reduction in the transcriptional activity of β-catenin in the CRC cells.

Fig. 1.

DHTS significantly reduces nuclear β-catenin expression and transcriptional activity in CRC cells. Western blot showing the cytoplasmic and nuclear expressions of β-catenin in (A) DLD-1 and (B) HT55 cells after 48 h DHTS treatments. (C) Western blot showing the protein expressions of β-catenin, LEF-1 and TCF-4 in β-catenin-overexpressed DLD-1 cells (DLD-1-β-catenin) and DLD-1 cells after 48 h LiCl treatments (50mM). (D) Western blot showing the protein expressions of β-catenin, LEF-1 and TCF-4 in β-catenin-overexpressed HT55-1 cells (HT55-1-β-catenin) and HT55 cells after 48 h LiCl treatments (50mM). Western blot showing the protein expression of LEF-1 and TCF-4 in (E) DLD-1 cells, (F) HT55 cells, (G) β-catenin-overexpressed DLD-1 cells (DLD-1-β-catenin) and (H) β-catenin-overexpressed HT55cells (HT55-1-β-catenin) after 48 h DHTS treatments. TOP/FOP luciferase reporter assays in (I) DLD-1 cells treated with DHTS at 1.6 µM and (J) HT55 cells treated with DHTS at 2 µM. Shown is the mean ± SEM, n = 3 individual experiments, ^*p < 0.05, ^**p < 0.01 compared with control, ^ap<0.05, compared with Top-β-catenin control. DHTS, dihydrotanshinone I; DHTS-L, low dosage of dihydrotanshinone I (1.2 µM for DLD-1, 1 µM for HT55); DHTS-H, high dosage of dihydrotanshinone I (1.6 µM for DLD-1, 2 µM for HT55)

DHTS physically binds to β-catenin protein

Next, we investigated the mechanism underlying how DHTS reduced the nuclear β-catenin in CRC. We found that DHTS treatments did not affect β-catenin mRNA level (Fig. S3A and S3B). It is known that GSK3β phosphorylates β-catenin in its N-terminal domain that leads to the degradation of β-catenin via the ubiquitin/proteasome pathway. Our data showed that DHTS treatments did not affect the total GSK3β and activated GSK3β levels in DLD-1 cells (Fig. S3C) and HT55 cells (Fig. S3D). The treatments also did not affect the cytoplasmic β-catenin expressions in the absence or presence of cell-permeable proteasome inhibitor MG132 (Fig. S3E-H), suggesting that DHTS does not affect β-catenin stability. Although expressions of nuclear pore complexes (NPCs) will affect the nuclear translocation of β-catenin [20], DHTS treatments did not affect the expressions of the critical NPCs including Ran BP-2, nucleoporin p62 and Nup98 in DLD-1 cells (Fig. S3I) and HT55 cells (Fig. S3J).

Interestingly, molecular docking analysis suggests that DHTS physically binds to β-catenin with binding energy of -6.7 kcal/mol. The three-dimensional ribbon model showed that DHTS occupied a hydrophobic pocket consisting of residues Ser-411, it also formed a hydrogen bond to Ser411 with a bond length [BL]of 2.89 Å in β-catenin (Fig. 2A-C). The RMSD (Root Mean Square Deviation) measures the average distance between the atoms of a docked ligand and the atoms of a reference structure. As shown in Fig. 2D, the DHTS-β-catenin interaction stabilized over time, with the RMSD fluctuated during the first half of the molecular dynamic simulation and was stabilized after 40 ns. To further suggest DHTS physically binds to β-catenin protein, we performed real time biolayer interferometry (BLI). Six different concentrations of DHTS ranging from 6.25 µM to 200 µM were tested. BLI showed that the binding signals increased as DHTS concentration increased, and the DHTS-β-catenin protein complex was stable (Fig. 2E and F). The dissociation (KD) was 1.46E-05 M, with association rate constant (kon) of 6.90E + 03 s^− 1 and dissociation rate constant (kdis) of 1.00E-01 s^− 1. While biophysical studies suggest that DHTS binds to β-catenin, it is still uncertain if this binding occurs in CRC cells. Therefore, we performed cellular thermal shift assay (CETSA), in which the ligand binding will lead to thermal stabilization of the target protein. This stabilization will be reflected in changes to the protein’s unfolding behavior and distinct melting curves when subjected to elevated temperatures. We found that DHTS induced a large thermal shift of β-catenin in CRC cells, as indicated in the melting curves (Fig. 2G and H). Based on our data, we strongly suggests that DHTS physically binds to β-catenin protein, which may affect its nuclear translocation in CRC cells.

Fig. 2.

DHTS physically binds to β-catenin protein. (A-C) The three-dimensional ribbon model of the DHTS-β-catenin complex and (D) the molecular simulation showing a stable binding between DHTS and β-catenin. (E) BLI sensorgrams showing the interactions between DHTS and β-catenin were measured on an Octet Red 96 system, with association and dissociation for 120s each. (F) Curves (red) fitting of the association and dissociation responses. CETSA analysis of the binding of DHTS to β-catenin in (G) DLD-1 cells and (H) HT55 cells. DHTS induces a large thermal shift (∆Tm) of β-catenin as indicated in the melting curve. Shown is the mean ± SEM, n = 3 individual experiments, ^*p < 0.05, ^**p < 0.01 compared with control. DHTS, dihydrotanshinone I

A point mutation at Ser411 on β-catenin disrupts the binding of DHTS and abolishes the inhibitory effect of DHTS on the nuclear expression of β-catenin

Results in the molecular docking analysis suggest that DHTS physically binds to Ser411 on β-catenin, which is within the armadillo repeat of the β-catenin protein (Fig. 3A). The armadillo repeat is a key region of β-catenin, characterized by its helix-loop-helix structure. This repeat plays a crucial role in facilitating the interaction between β-catenin and its ligands [21]. We constructed a mutated form of β-catenin (S411A) (Fig. 3B) and overexpressed it in HEK293 cells (Fig. 3C). Interestingly, data in the CESTA study suggests that DHTS does not bind to the mutated β-catenin S411A (Fig. 3D). We also overexpressed the mutated β-catenin S411A in DLD-1 cells (Fig. 3E) and treated these cells with DHTS. We found that the treatments failed to reduce the nuclear expression of β-catenin (Fig. 3F). The data strongly suggest that DHTS binds to β-catenin at S411 that impedes its nuclear translocation in CRC cells.

Fig. 3.

A point mutation at Ser411 on β-catenin disrupts the binding of DHTS and abolishes the inhibitory effect of DHTS on the nuclear expression of β-catenin. (A) A schematic diagram showing the armadillo repeat of the β-catenin protein. (B) Sequence of the β-catenin S411A mutation construct. Western blot showing the expressions of β-catenin in (C) HEK293 cells and (E) DLD-1 cells after β-catenin or β-catenin S411A overexpression. CETSA analysis of the binding of DHTS (24 µM) to β-catenin in (D) β-catenin-overexpressed HEK293 and β-catenin S411A-overexpressed HEK293 cells. (F) Western blot showing the expressions of the cytoplasmic and nuclear β-catenin in the β-catenin S411A-overexpressed DLD-1 cells after 48 h DHTS treatments. Shown is the mean ± SEM, n = 3 individual experiments, ^*p < 0.05 compared to control. DHTS, dihydrotanshinone I; DHTS-L, low dosage of dihydrotanshinone I (1.2 µM); DHTS-H, high dosage of dihydrotanshinone I (1.6 µM)

To examine whether DHTS changed β-catenin protein structure and hence affected its function and nuclear translocation, we performed circular dichroism (CD) analysis. Our data showed that the secondary structure of β-catenin contained 35.4% α-helix, 11.1% β-strand and 9.7% β-turns, with 43.8% unordered. While the corresponding secondary structures of β-catenin after binding with DHTS were 30.7% α-helix, 13.6% β-strand, 14.6% β-turns and 41.1% unordered, respectively (Fig. S4). The results suggest that the binding of DHTS to β-catenin does not significantly affect the secondary structure and protein folding of β-catenin.

CD36 is a β-catenin downstream signaling molecule in CRC

Given DHTS reduces the nuclear β-catenin and its transcriptional activity, we next explored the downstream signaling molecule of β-catenin that mediated the growth inhibitory effect of DHTS in CRC. Among the candidates we examined, we found that CD36 expressions were significantly increased if we overexpressed β-catenin in CRC cells (Fig. 4A). CD36 is a transmembrane fatty acid translocase, it has been implicated in cancer progression [22]. In CRC, CD36 promotes proliferation by activating Akt and survivin [23], and increases cancer metastasis by increasing matrix metalloproteinase expressions [24]. We performed staining with carcinoma tissue microarray which contained 60 cases of tumor tissues and 9 cases of normal tissues. We also found that CD36 expressions were significantly higher in human CRC samples compared to normal tissues (Fig. 4B). A high CD36 expression was also associated with low survival probability of the CRC patients (Fig. 4C). These findings further suggest that elevated CD36 expression is associated with CRC development.

Fig. 4.

CD36 is a β-catenin downstream signaling molecule in CRC. (A) Western blot showing the expression of β-catenin and CD36 in β-catenin-overexpressed DLD-1 cells (DLD-1- β-catenin) and control cells. (B) Immunohistochemical (IHC) staining of CD36 in the human colon cancer tissue and normal colon tissues (US Biomax #CO702b). (C) Correlation between the survival probability and CD36 expressions in CRC patients. Western blot showing the protein expressions of CD36 in (D) DLD-1 cells, (E) HT55 cells and (F) β-catenin-overexpressed DLD-1 cells (DLD-1- β-catenin) after 48 h DHTS treatments. The relative intracellular ATP levels in (G) DLD-1, (H) HT55, (I) CD36-overexpressed DLD-1 cells (DLD-1-CD36) and (J) β-catenin overexpressed DLD-1 cells (DLD-1-β-catenin) after 48 h DHTS treatments. Shown is the mean ± SEM, n = 3 individual experiments, ^*p < 0.05, ^**p < 0.01 compared to control. DHTS, dihydrotanshinone I; DHTS-L, low dosage of dihydrotanshinone I (1.2 µM for DLD-1, 1 µM for HT55); DHTS-H, high dosage of dihydrotanshinone I (1.6 µM for DLD-1, 2 µM for HT55)

If CD36 is a downstream signaling molecule of β-catenin, a reduction of β-catenin following DHTS treatments should reduce CD36 expression in CRC. Indeed, we found that DHTS treatments significantly reduced CD36 expressions in a dose-dependent manner (Fig. 4D and E), which was reversed when β-catenin was overexpressed in the cells (Fig. 4F). To examine the consequences of reducing CD36 in CRC, we examined the ATP levels in the cells because CD36 has been implicated in lipid metabolism [25]. As shown in Fig. 4G-H, DHTS treatments significantly reduced ATP levels in CRC cells in a dose-dependent manner, which was reversed when we overexpressed either CD36 (Fig. 4I) or β-catenin in the cells (Fig. 4J). The data strongly suggest that CD36 is a downstream signaling molecule of β-catenin, and that the β-catenin/CD36 axis mediates the reduction of ATP levels in CRC cells following DHTS treatments.

β-catenin/CD36 axis mediates the anti-CRC effect of the DHTS-loaded PLGA-PEG nanoparticles in vivo

DHTS has poor solubility and biocompatibility [26]. Therefore, we prepared DHTS-loaded PLGA- PEG nanoparticles (DHTS-NP) (Fig. 5A) before treating CRC-bearing mouse models. After DHTS-NP was administrated via intravenous injection, we performed HPLC to examine the DHTS levels in the tumors. Our data clearly showed that DHTS-NP treatments increased the levels of DHTS in tumor (Fig. S5). More importantly, the DHTS-NP treatments at either low or high dosages significantly reduced the tumor sizes and tumor weight in the mouse models (Fig. 5B-D). Compared with the control group, the tumor volume in the low-dosage DHTS-NP group was reduced by 67.8%, and the tumor volume in the high-dosage DHTS-NP group was reduced by 85.7%. The tumor weight in the low-dosage group was reduced by 62.8%, and the tumor weight in the high-dosage group was reduced by 88%. Moreover, DHTS-NP treatments demonstrated a more potent anti-CRC effect than the positive control drug 5-fluorouracil (5-FU) (Fig. 5B-D). In the 5-FU group, the tumor volume was reduced by only 59.8%, while the tumor weight was reduced by 64.7%. DHTS did not have apparent toxicity to the mice because the body weight in all the groups did not have apparent changes (Fig. S6A). The body weight in control group was 27.53 ± 1.99 g; low-dosage DHTS-NP group was 27.22 ± 0.61 g; high-dosage DHTS-NP group was 26.86 ± 1.18 g, and 5-FU group was 29.27 ± 1.15 g. The treatment also did not apparently affect the organ index of the mice (Fig. S6B).

Fig. 5.

β-catenin/CD36 axis mediates the anti-CRC effect of the DHTS-laden PLGA-co-PEG nanoparticles in vivo. (A) Flowchart showing the preparation of DHTS-laden PLGA-co-PEG nanoparticles (DHTS-NP). (B) The dissected tumor tissues of the CRC-bearing xenograft mouse models after DHTS-NP or 5-fluorouracil (5-FU) treatments. (C) Tumor volume and (D) tumor weight of the dissected tumors after DHTS-NP or 5Fu treatments. (E) Western blot showing the nuclear expressions of β-catenin in the xenograft tissues. Western blot showing the expressions of (F) cytoplasmic β-catenin, (G) LEF-1, TCF-4 and CD36 in the xenograft tissues. Shown is the mean ± SEM, n = 6–8 mice in each group, ^*p < 0.05, ^**p < 0.01 compared to control. DHTS, dihydrotanshinone I; DHTS-NP, DHTS-laden PLGA-co-PEG nanoparticles; DHTS-NP-L (15 mg/kg DHTS); DHTS-NP-H (20 mg/kg DHTS); 5Fu, 5-fluorouracil (10 mg/kg)

Our Western blot analysis also showed that DHTS-NP treatments significantly reduced the nuclear expressions of β-catenin (Fig. 5E) but not the cytoplasmic β-catenin level (Fig. 5F) in the tumors. The decreased expressions of TCF4, LEF1 and CD36 in the tumor tissues (Fig. 5H) following DHTS treatment were consistent with our findings in the CRC cell models. Taken together, the in vivo data strongly suggests that DHTS-NP treatment inhibits CRC tumor growth by reducing nuclear β-catenin and hence CD36 expression.

Discussion

Our study shows that DHTS does not affect the cytoplasmic β-catenin expression; however, it significantly reduces nuclear β-catenin expression and transcriptional activity in CRC by physically binding to the β-catenin at Ser411. Interestingly, a point mutation at Ser411 on β-catenin disrupts this binding, abolishing the inhibitory effect of DHTS on nuclear β-catenin expression, suggesting this binding impedes the nuclear translocation of β-catenin in CRC cells. Furthermore, this study is among the first to identify CD36 as a downstream signaling molecule of β-catenin in CRC. The inhibitory effect of DHTS on the β-catenin/CD36 axis and the subsequent reduction in ATP level suggest a novel therapeutic strategy for developing DHTS or DHTS-encapsulated nanoparticles as a potential agent for CRC treatments.

Approximately 90% of the CRC cases harbor mutations in the Wnt/β-catenin signaling pathway [4]. Aberrant activation of Wnt/β-catenin signaling drive CRC development by regulating cancer proliferation, apoptosis, autophagy, metastasis, inflammation and metabolism [27]. β-catenin, acting as a coactivator, binds to the TCF4/LEF1 transcription factor complex, converting TCF4 into a transcriptional activator that initiates gene transcription. A study has been done to identify the β-catenin/TCF4 target genes in CRC using cell model HCT116 and ChIP-Seq technique. This study has identified a total of 988 genes that contain β-catenin enriched regions [28]; however, CD36 is not identified as one of the target genes. Our study is among the first to report CD36 as a downstream signaling molecule of β-catenin, via a mechanism that has not been reported. Other studies have shown that nuclear receptor peroxisome proliferator activated receptor gamma (PPAR-γ) enhances CD36 gene transcription [29, 30]. Nevertheless, there is a complex relationship between PPAR-γ and β-catenin. Stabilization of β-catenin increases PPARγ expression [31], while PPARγ increases β-catenin degradation [32] This feedback mechanism between β-catenin and PPAR-γ is unlikely to underlie the increased CD36 level in β-catenin-overexpressed cells. Nevertheless, the promoter region of CD36 also consists of STAT binding GAS elements that can be regulated by a number of STAT transcription factors [33–35]. Among these STAT transcription factors, STAT3 expression and activity can be increased by β-catenin as reported in lymphoma [36]. Investigating whether STAT3 mediates the increased CD36 levels in β-catenin–overexpressing CRC cells will provide valuable insights for developing therapeutic strategies targeting β-catenin and CD36.

CD36 plays an important role in CRC development by facilitating fatty acid uptake that supports the aggressive growth of cancer [22]. Interestingly, in CRC, CD36 inhibits β-catenin-mediated glycolysis [37], thereby increasing a dependence on fatty acids for energy production. Fatty acids serve as substrate for mitochondrial β-oxidation and oxidative phosphorylation, hence increases ATP production. During periods of energy demand such as fasting or exercise, increased lipolysis from adipocytes and very low-density lipoprotein secretion by the liver deliver fatty acids to tissues, particularly skeletal muscle and heart. Interestingly, it has been shown that CD36 is essential for long chain fatty acid uptake in heart and skeletal muscle [38]. Indeed, it has been recognized that CD36 plays an important role in lipid metabolism [25, 39]. Under CD36 deficiency, cells will shift to glucose for energy. However, this would limit ATP generation compared with the fatty acid oxidation pathway. The importance of CD36 in relation to ATP production is also reported in cancer cells. Study shows that cancer with high CD36 expression imports lipids to fuel oxidative phosphorylation; while blocking CD36 reduces fatty acid oxidation flux and cellular ATP, hence reducing the cancer survival even in lipid‑rich environments [40].

Although both fatty acid synthase (FASN) and CD36 are involved in lipid metabolism, drugs targeting FASN for CRC treatment may not be an effective therapeutic strategy because inhibition of FASN selectively upregulates CD36 translocation to the plasma membrane of the CRC cells [23] to increase fatty acid uptake and restore the cellular fatty acid levels. Then again, FASN inhibitors for CRC are still in preclinical stage. Our study has clearly demonstrated that DHTS treatments significantly reduce both the CD36 and nuclear β-catenin expressions in CRC and the ATP levels in CRC. Although DHTS has poor solubility, various nanoparticles have been designed to encapsulate the compound [26]. Moreover, nanoparticle therapies have been clinically approved [41]. In our study, we have also shown that the DHTS-loaded PLGA-PEG nanoparticles exhibit a strong anti-CRC effect in the mouse model, which is even more potent than the frontline drug fluorouracil, 5-Fu. DHTS, as a natural compound, possesses great clinical translation for CRC treatment by targeting the β-catenin/CD36 axis.

Many strategies have been proposed to inhibit the Wnt/β-catenin signaling pathway. These strategies include targeting the cell membrane to inhibit Wnt ligands, reducing β-catenin stability in the cytoplasm, or modulating the transcription of β-catenin-targeted genes in the nucleus. Targeting upstream molecules in the Wnt/β-catenin pathway for cancer therapy may be disadvantageous. Cancer-associated mutations often lead to upregulation of downstream effectors and crosstalk with other pathways, rendering upstream-targeted therapies less effective and potentially promoting resistance. Antibody-based therapy that targets Wnt ligands and prevent Wnt-FZD interactions [42] and the small molecule-based therapies such as porcupine inhibitors (PORCNi) [43, 44] have not been approved by FDA for the treatment of CRC. Modulating the transcriptional activity of β-catenin in the nucleus is another strategy to treat cancer. There are several key binding regions on β-catenin that facilitate its interaction with TCF4, including β-catenin residues N426, K435, R469, H470 and K508 which form a concave pocket to interact with TCF4 [45]; β-catenin residues K312 and K345 bind to TCF4 residues E24 and E29 [46]; or the residues H578 and R582 of β-catenin interact with TCF4 residue D11 [47]; while the hydrophobic region surrounding β-catenin residues F253, F293, and I296 interact with TCF4 through hydrophobic interactions [48]. Since these binding sites of β-catenin for TCF4 overlaps with that of Axin, peptide inhibitors have been developed to interrupt the binding between β-catenin, and TCF4 and Axin [49, 50]. However, these peptides need to be structurally optimized to enhance their cellular uptake. Besides, biochemical and biophysical studies have been done to identify compounds that interrupt the transcriptional complexes. For example, ICG-001 [51] and its second-generation prodrug derivatives PRI-724 [52], CWP232228 [53] and CWP232291 [54]. These compounds disrupt the CREB-binding protein (CBP)/β-catenin interaction by binding to the general coactivator CBP. Therefore, the mechanism may not be Wnt pathway-specific, given CBP regulates many transcriptional activities [55]. Of these compounds, CWP232291 entered a Phase I clinical trial in year 2015 for patients with relapsed or refractory acute myeloid leukemia and myelodysplastic syndromes (ClinicalTrials.gov NCT01398462); while PRI-724 was tested for liver cirrhosis that resulted in liver injury, which was reported as an adverse event (ClinicalTrials.gov NCT02195440). To date, none of these compounds have received FDA approval. There may have a drawback to targeting β-catenin/TCF interaction. Some studies have shown that targeting β-catenin/TCF interaction may activate Wnt/β-catenin signaling [56, 57]. Furthermore, the binding of β-catenin to TCF4 and APC may overlap with its interaction with BCL9 and E-cadherin. Therefore, small molecules disrupting the β-catenin/TCF4 interaction should exhibit high selectivity. Table S2 shows the comparison of DHTS with different β-catenin inhibitors.

Targeting the Wnt/β-catenin pathway with compounds that directly bind and inhibit β-catenin may be a more effective cancer treatment strategy. In the past, several compounds were suggested to directly bind to β-catenin including PNU-74,654, which binds to β-catenin residues K435 and R469 [58]; and MSAB binds to second armadillo repeat of β-catenin at residue K301-Y670 [59]. However, many of these compounds were identified only though biochemical or biophysical studies, and their anti-cancer effects, particularly in CRC, have not been examined. Moreover, compounds like PNU-74,654 contain PAINS moieties, such as acyl hydrozones. These reactive functional groups can hinder modification of the compound to improve its potency, selectivity, and physicochemical properties [60, 61].

Our study has revealed that a mutation at S411 of β-catenin disrupts the binding of DHTS, indicating that S411 is a ligandable site. Notably, DHTS treatment does not reduce the nuclear expression of β-catenin in CRC cells that overexpress the S411A mutant. Since DHTS treatment does not influence cytoplasmic β-catenin levels or the essential nuclear pore complex protein, we can reasonably postulate that DHTS binding to β-catenin inhibits its nuclear translocation. Indeed, Ser411 lies within the armadillo repeat core which is a surface for the binding of TCF/LEF and to contact nuclear transport machinery either directly with Phe-Gly repeat motifs in nucleoporins or via import receptors or co-chaperones. DHTS may have occupied this region and hence affected the nuclear transport. Further study is needed to reveal the underlying mechanism of action. Strategies to inhibit the nuclear translocation of β-catenin have been suggested in other cancer types. For instance, in nasopharyngeal carcinoma, CGP57380, an inhibitor of MNK (MAP kinase interacting serine/threonine kinase), has been shown to inhibit β-catenin nuclear translocation by modulating Akt protein kinase activity [62]. However, CGP57380 also increases cytoplasmic β-catenin levels in the cancer [62], which may counteract its effectiveness in inhibiting β-catenin nuclear translocation. Our study not only identifies a novel ligandable site S411 on β-catenin but also proposes a mechanism by which DHTS inhibits the transcriptional activity of β-catenin in CRC and effectively inhibits CRC growth. However, for the clinical translation of DHTS, the co-crystal structure of DHTS with β-catenin has to be examined to optimize the compound to be an even more potent β-catenin inhibitor.

Nevertheless, this study has limitations. Our data clearly show that binding of DHTS to β-catenin at Ser411 impede β-catenin nuclear translocation. However, isothermal titration calorimetry (ITC), surface plasmon resonance (SPR) and X-ray crystallography can be done to further validate the binding between DHTS and β-catenin. We have identified CD36 is a downstream target of β-catenin. In future, CHIP assay can be done. It is expected that β-catenin and TCF/LEF should bind to the regulatory element such as promoter or enhancers of the CD36; while mutation of the β-catenin and TCF/LEF motif-containing CD36 promoter or enhancer should abolish the binding. More importantly, RNA-sequencing or proteomics can help to suggest whether DHTS treatments would induce any downregulation or upregulation of the signaling molecules that are unrelated to β-catenin signaling pathway. This data can demonstrate the specificity of the β-catenin targeting property of DHTS and greatly enhance its translational potential into a β-catenin-specific inhibitor.

Conclusion

In conclusion, our findings highlight the potential of DHTS as a therapeutic agent for CRC. By physically binding at Ser411, DHTS effectively reduces nuclear β-catenin levels. Moreover, DHTS-loaded PLGA-PEG nanoparticles significantly inhibit CRC growth via the β-catenin/CD36 axis in vitro and in vivo. Our study not only underscores the importance of Ser411 in β-catenin but also opens avenues for developing DHTS or DHTS-loaded nanoparticles as a β-catenin targeted therapies for CRC.

Materials and methods

Reagents and chemicals

DHTS (catalog #D0947, Sigma-Aldrich) was dissolved in DMSO to a maximum concentration of 12.5 mM. Primary antibodies were obtained from multiple suppliers: β-catenin (catalog #BD610153) from BD Biosciences; TCF-4 (catalog #sc-166699), LEF-1 (catalog #sc-374522), Ran BP-2 (catalog #sc74518), nucleoporin p62 (catalog #sc48373), Nup98 (catalog #sc74578), Nup153 (catalog #sc101544), and β-actin (catalog #12262) from Santa Cruz Biotechnology; and phospho-β-catenin (Ser33/37/Thr41) (catalog #9561), β-tubulin (catalog #2146), lamin B1 (catalog #12586), phospho-GSK-3β (Ser9) (catalog #9336), and GSK-3β (catalog #9315) from Cell Signaling Technology.

Cell lines and culture

CRC cell lines DLD-1, HT55, SW480, HCT116, SW620, HT29, HCT15, Caco-2, and the mouse cell line CT26 were obtained from the American Type Culture Collection (ATCC). All cell lines were tested for mycoplasma contamination prior to use. Human CRC cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS; Life Technologies Ltd) and 1% penicillin-streptomycin. Mouse CRC cells were cultured in RPMI 1640 Medium supplemented with 10% FBS (Life Technologies Ltd) and 1% penicillin-streptomycin. All cell cultures were maintained at 37 °C in a humidified incubator with 5% CO2. The culture medium was refreshed every three days, and cell passaging was performed using 0.05% trypsin/ethylenediaminetetraacetic acid (EDTA).

MTT assay

To assess the cytotoxicity of DHTS on CRC cells, MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay was employed. DHTS was dissolved in dimethyl sulfoxide (DMSO) and diluted into a series of concentrations with culture medium. CRC cells in the logarithmic growth phase were seeded in 96-well plates at a density of 4,000 cells per well and incubated at 37 °C in a 5% CO₂ atmosphere for 24 h. Subsequently, the cancer cells were treated with DHTS at different concentrations for 48 h. After the treatment, 20 µL of MTT solution was added to each well, and the plates were incubated at 37 °C for 4 h. Following incubation, the supernatant was aspirated, and 150 µL of DMSO was added to each well to dissolve the formazan crystals. The plates were then placed on a shaker for 5 min to ensure complete dissolution. The absorbance at 570 nm was measured using a microplate reader. Cell viability and half maximal inhibitory concentration (IC₅₀) values were determined using GraphPad Prism 8 software.

Cytoplasmic and nuclear protein extraction

Nuclear and cytoplasmic proteins from cell lines and tumors were extracted according to the manufacturer’s protocol (NE-PER™ Nuclear and Cytoplasmic Extraction Reagents, catalog #78833, Thermo Fisher Scientific). Briefly, CRC cells were harvested using trypsin-EDTA and then washed with PBS. The cytoplasmic proteins were extracted by treating the cells with ice-cold Cytoplasmic Extraction Reagent I (CER I) followed by Cytoplasmic Extraction Reagent II (CER II). After centrifugation at maximum speed (~ 16,000 × g) for 5 min in a microcentrifuge, the supernatant containing the cytoplasmic proteins was collected. The remaining pellet was resuspended in ice-cold Nuclear Extraction Reagent (NER). The mixture was vortexed at the highest setting for 15 s every 10 min, for a total of 40 min, to extract the nuclear proteins. For tumor tissue protein extraction, approximately 50 mg of tumor tissue per mouse was used.

Western blot analysis

Western blotting was employed to ascertain the levels of specific proteins of interest. Protein concentrations were quantified using a BCA protein assay. Subsequently, the proteins were subjected to electrophoresis using either 8% or 10% SDS-PAGE and then electrotransferred to a PVDF membrane. The membrane was blocked with a 5% solution of blotting-grade blocker in PBST for one hour, followed by an overnight incubation at 4 °C with the primary antibody. After incubation, the membrane was washed three times with PBST buffer and then incubated with the secondary antibody for one hour at ambient temperature. Finally, the proteins were visualized using an enhanced chemiluminescence (ECL) detection system (Thermo Fisher Scientific).

TOP/FOP luciferase reporter assay

To assess the transcriptional activity of β-catenin in CRC cells, we employed the TOP/FOP reporter system using the Dual-Luciferase^® Reporter Assay kit (Promega, USA). CRC cells were transiently transfected with 500 ng of either the β-catenin-responsive firefly luciferase reporter plasmid TopFlash (catalog #35489, Addgene) or the negative control FopFlash (catalog #35490, Addgene) along with the p-RL-TK reporter plasmid using Lipofectamine 3000 Reagent (Thermo Fisher Scientific). For β-catenin overexpression experiments, CRC cells were transiently co-transfected with 250 ng of β-catenin (catalog #16828, Addgene) and 250 ng of either the TopFlash or FopFlash reporter plasmid, along with the p-RL-TK reporter plasmid. Cells were harvested after 24 h in culture, followed by 48 h of treatment with dihydrotanshinone I. Both firefly and Renilla luciferase activities were measured according to the manufacturer’s instructions. The firefly luciferase activity was normalized against the Renilla luciferase activity and the fold increase in TOPFlash activity compared to FOPFlash is reported.

RNA extraction and real-time PCR (RT-PCR) reaction

Total RNA from CRC cells was extracted using Trizol reagent (Invitrogen) according to the manufacturer’s protocol. cDNA was synthesized in a total volume of 20 µL using a cDNA reverse transcription reaction kit (TAKARA). RT-PCR was performed with a PrimeScript™ RT reagent Kit (TAKARA) using a Viia7 RT-PCR machine (Thermo Fisher Scientific). All mRNA expression levels were normalized to that of GAPDH.

ATP measurement

ATP levels were measured using the Enhanced ATP Detection Kit (catalog number S0027, Beyotime Biotechnology, China) following the manufacturer’s instructions. DLD-1, HT55, β-catenin-overexpressing DLD-1 (DLD-1-β-catenin), and CD36-overexpressing DLD-1 (DLD-1-CD36) cells were treated with DHTS for 48 h at the following concentrations: DLD-1, DLD-1-β-catenin, and DLD-1-CD36 cells at 1.2 µM and 1.6 µM, and HT55 cells at 1 µM and 2 µM. Post-treatment, the cells were lysed using ATP lysis buffer and centrifuged at 12,000 × g for 5 min. The supernatant was then used for ATP quantification with an illuminometer. Additionally, the protein content in each sample was determined using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific).

Molecular docking simulation

The interaction of DHTS with the β-catenin receptor was investigated employing Autodock Vina version 1.1.2. The three-dimensional structure of the β-catenin complex was obtained from alphafold database (ID: P35222). DHTS’s 3D structure was created using ChemBioDraw Ultra 14.0 as well as ChemBio3D Ultra 14.0. The AutoDockTools 1.5.6 suite was used to create the necessary input files for docking.

Molecular dynamics

A molecular dynamics simulation was performed using the amber18 program. All simulations were run with the amber14SB force field. After energy minimization, velocities were rescaled only about every 100 simulation steps, whenever the average of the last 100 measured temperatures converged. After that, 100ns MD simulations were performed with a time step of 2 fs and the coordinates of the complexes were saved every 20 ps.

Biolayer interferometry (BLI) assay

The binding affinity of β-catenin with DHTS was determined using Super Streptavidin (SSA) biosensors in an Octet Red 96 instrument (ForteBio Inc., Menlo Park, CA, USA). SSA biosensors with immobilized biotinylated β-catenin were exposed to a range of DHTS concentrations (6.25 µM, 12.5 µM, 25 µM, 50 µM, 100 µM, and 200 µM) at 25 °C. The assays were conducted in black solid 96-well flat-bottom plates on a shaker set at 1,000 rpm. The association and dissociation of DHTS with β-catenin were measured for 4 min. Data was obtained using the ForteBio software Data Acquisition v7.0. Kinetic parameters and affinities were calculated from a non-linear global fit of the data between DHTS and β-catenin using Octet Data Analysis software version 7.0 (ForteBio).

Confocal imaging

DLD-1 cells were transfected with β-catenin-GFP (#71367, addgene) plasmids with Lipofectamine 3000 Reagent (Thermo Fisher Scientific) for 24 h. Then the DLD-1-β-catenin-GFP cells were treated with 0, 1.2 or 1.6 µM dihydrotanshinone I for 48 h. After the treatment, the cells were fixed by 4% paraformaldehyde for 15 min, and stained by 0.1 µg/mL 4’,6-diamidino-2-phenylindole (DAPI) for 5 min. The signals were detected by Confocal Laser Scanning Microscopy (Leica) under 120x magnification.

β-catenin S411A mutation construct

The Megaprimer method was used to perform point mutations (SYNBIO Technologies, China) on the β-catenin construct (Addgene 71367), and the serine 411 in the gene was mutated to alanine. The primers used for the mutation were β-catenin-1V-seqF: AGGTCTATATAAGCAGAGCT; β-catenin-1V-seqR: CCGGACACGCTGAACTTGTG; β-catenin − 1-seq2: TTCCAGACACGCTATCATGC; β-catenin − 1-seq3: GTGCTGATGATATAAATGTG; β-catenin-1F1: 5’-CTATCAAGATGATGCAGAACTTGCC ACACGTG-3’; β-catenin-1R1: 5’-CTGCACAGGTGACCACATTTATATCATCAG CACCCAGAAGCTGAACAAGAGTCCCA-3’; β-catenin-1F2: 5’- TGGGACTCTTGTTCAGC TTCTGGGTGCTGATGATATAAATGTGGTCACCTGTGCAG-3’; β-catenin-1R2: 5’-TAT CCTGATGTGCACGAACAAGCAACTGAACTAGT-3’.

Cellular thermal shift assay (CETSA)

DLD-1, HT55, HEK293-catenin, and HEK293-S411A cell lines were subjected to ten freeze-thaw cycles using liquid nitrogen. The resulting cell lysates were mixed with RIPA buffer and split into two portions. One portion was treated with DHTS at a concentration of 24 µM, while the other served as a control and was treated with DMSO. The samples were incubated at room temperature for 30 min, followed by heating at various temperatures: 55 °C to 85 °C for DLD-1, HEK293-catenin, and HEK293-S411A cell lines, and 70 °C to 88 °C for HT55 cells, for 15 min. The samples were then cooled at 4 °C for 3 min. Afterward, the samples were centrifuged at 12,000 g for 15 min at 4 °C to separate the supernatant. The collected supernatant was then subjected to SDS-PAGE and analyzed via Western blotting.

Overexpression of β-catenin, S411A or CD36 in CRC cells and HEK293 cells

Transient transfection assays were performed using Lipofectamine 3000 Reagent (Thermo Fisher Scientific) according to the manufacturer’s protocol. Human β-catenin construct (Addgene plasmid #71367) and human CD36 construct (Addgene plasmid #55011) were used for the overexpression studies.

Tissue microarray (TMA) sections and immunohistochemistry (IHC) analysis

Colon cancer tissue array of the human normal colon tissues and CRC tissues were obtained from US Biomax (CO702b). IHC analysis was performed with the paraffin-embedded tissue sections (4 μm) deparaffinized in 100% xylene and re-hydrated through a series of graded ethanol solutions. The TMA sections were incubated with a primary monoclonal rabbit anti-CD36 antibody (catalog number 14347, Cell Signaling Technology) in Tris-buffered saline (TBS) containing 1% bovine serum albumin (BSA) for 1 h. After washing with TBS, the sections were incubated with a horseradish peroxidase-conjugated anti-rabbit secondary antibody (Bio-Rad). The TMA sections were then examined under a light microscope.

Preparation of DHTS-laden PLGA-co-PEG nanoparticles

Copolymer poly(ethylene glycol)-block-poly(lactide-co-glycolide) (PLGA-co-PEG) was synthesized according to previously published literature. The molecular weight for PLGA block was 10 kDa and the molecular weight for PEG block was 5 kDa. The ratio between lactide to glycolide was 75:25. DHTS-laden PLGA-co-PEG nanoparticles (DHTS-NP) were prepared using an emulsion-solvent evaporation method as previously reported, with some modifications. Briefly, DHTS (3 mg) and PLGA-PEG polymer (30 mg) were dissolved in dichloromethane (DCM, 3 mL) to form the oil phase. This organic solution was then added to 30 mL of 0.5% polyvinyl alcohol (PVA) in an ice bath and sonicated with an ultrasonic homogenizer (SM-1000 A) for 3 min at 60% amplitude to obtain an oil-in-water (O/W) emulsion. The final solution was left under stirring for at least 4 h to evaporate the DCM. The resulting solution was centrifuged at 10,000 rpm for 10 min at 20 °C, washed twice with phosphate-buffered saline (PBS) to remove excess PVA and free Dihydrotanshinone I, and the appropriate amount of PBS was added to achieve the final concentration used in animal study. DHTS -free NPs were produced using a similar method without adding DHTS.

Circular dichroism (CD) analysis

CD spectra was acquired with a Chirascan circular dichroism spectrometer (Applied Photophysics Ltd., London, United Kingdom), a spectropolarimeter equipped with a Peltier. In the far- UV range of 195 to 260 nm, the CD spectra of β-catenin (Sino Biological, #11279-H20B, 0.35 mg/mL, dissolved in sterile water) in combination with DHTS (48 mM, dissolved in DMSO) were observed. The reaction system comprised 100 µL of β-catenin solution mixed with 0.1µL DHTS-48mM or 0.1µL DMSO, which reacted for 2 h at 4℃. Each spectrum displayed is an average of three scans. All of the CD data were expressed in the form of the ellipticity (mdeg). The secondary structure of proteins is calculated online at http://bestsel.elte.hu.

CRC-bearing mouse model

Balb/c nude mice were obtained from the Laboratory Animal Services Centre at the Chinese University of Hong Kong. The mice were housed in ventilated cages in the animal room at Hong Kong Baptist University, maintained on a 12-hour light-dark cycle, and provided with unlimited access to food and water. To establish the CRC-bearing xenograft mouse model, 6 weeks old male Balb/c nude mice were used. DLD-1 cells (4 × 10⁶) in 100 µL of PBS were subcutaneously inoculated into the right flank of the nude mice. When the tumor volumes reached approximately 50 mm³, the mice were randomly assigned into groups and treated with DHTS-loaded nanoparticles (15 mg/kg or 20 mg/kg) once daily for 20 days via the intraperitoneal route. 5-fluorouracil (5-FU) was administered intravenously at a dose of 10 mg/kg. Tumor volume was measured using a caliper and calculated with the following formula: V = (L×W²)/2, where L and W represent the length and width of the tumor, respectively. After the treatment period, the mice were sacrificed, and the tumors were dissected and weighed.

LC/MS-based DHTS detection in tumors

The xenograft tissues (20 mg) from each mouse were homogenized by sonication in 1 mL of methanol. After centrifugation at 15,000 rpm for 30 min at 20 °C, the supernatant was collected for liquid chromatography-mass spectrometry (LC/MS) analysis. The LC/MS-based lipidomics analysis and data processing were performed using an Agilent 6460 QQQ LC/MS mass spectrometer (Agilent Technologies), which was connected to an Agilent 1290 Infinity UHPLC via an electrospray ionization (ESI) ion source for the detection of Dihydrotanshinone I. Chromatographic analysis was conducted on an Agilent 1290 Infinity UHPLC system, and separation was achieved using a Waters Acquity BEH C₁₈ column (2.1 × 50 mm). The mobile phases consisted of 0.1% formic acid in water (A) and 0.1% formic acid in acetonitrile (B). A gradient elution was adopted as follows: 0–8 min, 15%-60% B; 8–10 min, 60%-100% B; 10–13 min, 100% B; 13–13.1 min, 100%-15% B; 13.1–15 min, 15% B. The flow rate was 0.35 mL/min. All analytes were eluted within 15 min. The column temperature was maintained at 25 °C, and the injection volume was 2 µL. Mass spectrometric detection was performed on a triple quadrupole 6460 mass spectrometer (Agilent, USA) equipped with an ESI source. ESI-MS/MS was operated in positive mode for the analysis of Dihydrotanshinone I. The MS conditions were set as follows: capillary voltage at 3500 V (ESI⁺), nitrogen as the nebulizer gas at 40 psi, drying gas flow at 6 L/min, and drying gas temperature at 300 °C. Data acquisition and analysis were carried out using Agilent MassHunter Software (version B.06.00). The precursor ion (m/z) for Dihydrotanshinone I was 279.1, and the product ion was m/z 233.

Statistical analysis

We used SPSS 20.0 software to perform statistical analysis of the data, one-way analysis of variance (ANOVA) to examine the variance within each group and the significant difference between groups with *p < 0.05 and **p < 0.01. The data are shown as mean ± SEM, n = 3 independent experiments.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

BLI

Biolayer interferometry

CRC

Colorectal cancer

CSS

Cancer-specific survival

CETSA

Cellular thermal shift assay

CD

Circular dichroism

CBP

CREB-binding protein

DFS

Disease-free survival

DHTS-NP

DHTS-laden PLGA-co-PEG nanoparticles

FASN

Fatty acid synthase

HuR

Human antigen R

ITC

Isothermal titration calorimetry

LEF1

Lymphoid enhancer-binding factor-1

LiCl

Lithium chloride

MNK

MAP kinase interacting serine/threonine kinase

NPCs

Nuclear pore complexes

NER

Nuclear extraction reagent

OS

Overall survival

PORCNi

Porcupine inhibitors

PPAR-γ

Receptor peroxisome proliferator activated receptor gamma

RMSD

Root mean square deviation

SPR

Surface plasmon resonance

TCF4

T cell factor-4

5-FU

5- fluorouracil

DHTS

15,16-Dihydrotanshinone I

Author contributions

Hongjie Zhang, Hoi Leong Xavier Wong, Ying Ji, Tao Su, Hiu Yee Kwan contributed conception and design of the study. Minting Chen, Baisen Chen, Qianyi He, Shilin Xiao, Baoting Li, Yun Ye performed the experiments and interpreted the data. Minting Chen, Ying Ji, Tao Su, Hiu Yee Kwan wrote the manuscript. Hongjie Zhang, Hoi Leong Xavier Wong, Ying Ji, Tao Su, Hiu Yee Kwan revised the manuscript. All authors have given approval to the final version of the manuscript.

Funding

This work was partially supported by Natural Science Foundation of Guangdong Province (2023A1515011811, 2025A1515010488), Health and Medical Research Fund (08193596), Initial Grant for Faculty Niche Research Areas (RC-FNRA-IG/23–24/SCM/01) and Seed Funding for Collaborative Research Grants (RC-SFCRG/23–24/SCM/02) to HYK; Guangzhou Basic and Applied Basic Research Foundation (2025A04J5398), Young Pearl River Scholar of Guangdong Province, the Characteristic Innovation Projects of Universities in Guangdong Province (2023KTSCX023), and Administration of Traditional Chinese Medicine of Guangdong Province (20241074) to TS.

Data availability

The datasets generated and analyzed during the current study are available from the corresponding author upon request.

Declarations

Ethical approval

Hong Kong Baptist University Research Ethics Committee approved all animal care and experiments conducted in this study (#HASC/17–18/0456). The animal studies were approval by the Committees of Animal Ethics and Experimental Safety of Hong Kong Baptist University and the Department of Health under Hong Kong legislation.

Consent for publication

We promise to agree to publish our article in this magazine.

Competing interests

The authors declare no competing interests.